Aus.Social's Honking Admin

ROUTER UPGRADE COMPLETED

edit: and the "you were hacked" flagged is no. "The Machine God watches over us"

https://forum.mikrotik.com/t/important-security-update/272851

CRITICAL SECURITY UPDATE:

I need to patch our router (in Melbourne from Adelaide)....

If goes down and doesn't return, I'm going to fly over and rebuild the router in person... so it will be down until like 9pm worse case.

but should be down for like 30 seconds

https://forum.mikrotik.com/t/7-24-2-stable-is-released/272800

A.S Updated to latest Mastodon v4.7.2

https://github.com/mastodon/mastodon/releases/tag/v4.7.2

A.S has been updated to the latest release 4.7.1

https://github.com/mastodon/mastodon/releases/tag/v4.7.1

RE: @aussocialadmin@aus.social

We're back at zero.

ROUTER UPGRADE STARTING

A.S is a little sad.. it ran out of space and now it needs to process the last few hours of failed jobs

Everything should just return to normal once those queues empty

A.S Updated to latest stable v4.7.0

https://github.com/mastodon/mastodon/releases/tag/v4.7.0

RE: @aussocialadmin@aus.social

we're back

Reminder that I am always looking for "cheap" rackmounted servers!

If you know anybody decommissioning or selling semi recent gear. The @PostersUnionAU@aus.social has a modest budget to buy some good gear.

Please toot at me if you know anybody with a large looking paperweight that might be better suited by our community!

Those 16U of unused rackspace in Melbourne craves kit!

The @PostersUnionAU@aus.social family is growing soon!

Currently working with @nicdex@techhub.social (Previous administrator) and @Raccoon@techhub.social (Moderation) to transfer ownership of the popular https://techhub.social/ Mastodon Instance.

This is where the Patreon money is going to be able to support other Fediverse administrators who are no-longer able to support the operational time and pressure... which there is a lot.

I'm proud to be able to keep this large mastodon instance online for the future.

@sjoerd@aus.social @evilchili@linernotes.club :hiero:

Updated the BIOS and Kernel on both proxmox nodes and you didn't even notice! Migrated all of the VMs to one host at a time between patching allows for zero downtime! :ablobwobble:

These were mostly security patches but might have performance benefits.

RE: @aussocialadmin@aus.social

Everything is back to normal. This wasn't actually the DB fault... but the new traefik HTTP/3 wasn't enabling compression so the traffic to your browser was just taking longer to reach us.

The performance should be slightly better actually now I've diagnosed that issue and the traffic is compressed using zstd (fancy).

Happy Mother’s Day!

A.S is back online with a fancy new DB! (should be nice and snappy)

The server queue is going to take a while to catchup from the last 4 hours of downtime, but by the time people read it.. it would be completed by the morning :Koala:

The DB has decided to be slow.

Hasn't lost any data but considering I migrated to this new DB last month, this is just annoyting..... I'll investigate the issue tomorrow.

A.S Updated to the latest Mastodon code with the new profile design.

https://blog.joinmastodon.org/2026/03/a-redesign-for-profiles/

Update: Please send feedback to @MastodonEngineering@mastodon.social if you're up for that.

Test. Ok. the new S3 bucket is live and the CDN is working.

Now to recover the broken images from today.

A.S - S3 Storage is migrating to the new https://fedicloud.au "FediStorage".

When this post is about 10-30 minutes old... it should be completed

(A few images will fail to load for a few hours until I complete the final replication passes)

When this toot is 15 minutes old, I'm starting some maintenance to migrate the DB to Ubuntu 26.04 + Postgres 18.

ETA: 4 Hours